ixeption

4 ms·

1. 1.▲
   by ixeption 4mo ago

   I really think that this blog is part of the IPO story from Anthropic or OpenAI. Because the reality is: A generalist can simply not guide the LLMs through the system nor can he/she verify the output. LLMs are force-multipliers and if
2. 2.▲
   Claude Opus 4.7 leaks system prompt randomly(old.reddit.com)

   2 points by ixeption 5mo ago|0 comments
3. 3.▲
   Cognitive Debt: How AI-Generated Code Introduces a More Dangerous Kind of Debt(substack.com)

   2 points by ixeption 6mo ago|0 comments
4. 4.▲
   Agentic Engineering: The good, the bad, the ugly(substack.com)

   2 points by ixeption 7mo ago|0 comments